cDox

Series: Big Tech Alternatives

tl;dr: cDox is a new Canadian cloud document service, with a promising core but clearly some work to do.

In a previous post, I briefly added an update that cDox was a new contender in cloud document editing. It is Canadian hosted and promises not to train any AI on your data, so that's two pretty strong arguments out of the gate.

This is not a product category that I really need a lot, but I decided to do some introductory exploration. Note that this is an extremely new work in progress, so any criticism is quite possibly something that will get resolved within a reasonable amount of time.

A Basic Editor

It is clean and friendly to use. There's an easy dashboard showing your documents and the ability to create more as well as a page with help for how to get started.

For the most part, it is not trying to do anything that you don't expect from a simple document editor, which I am considering a good thing: he's getting the basics right before likely adding more over time.

One feature already present that did stand out to me, though, is a built-in export to Markdown. There's a very specific scenario where that is useful to me. If I am giving a presentation I currently write it in Word so I can print it with good formatting and it is a bit easier to move sections in a long document. Then afterward, I want it in Markdown to post on one of my eleventy sites. Word does not have an easy Markdown export built in, as far as I know, so I end up copy and pasting and then having to clean it all up. This cDox might give me the best of both worlds. Is that enough of a reason to change at least those small workflows? Maybe not, but it's definitely something.

Accessibility Issues

I said it is clean and easy to use, but by that I really mean with a big asterisk: for sighted users. I have some doubts about accessibility being properly considered in the design.

I quickly noticed that the heading level of the site is bad. There's no h1. There's something that looks like an h1, the page title text. Then there are h2's depending on the page. But there that page title was actually a button, which is not right. That should be a linked heading, not a button. On another page, there was an H1 but then immediately followed by an h3 before the rest of the document made sense again.

This often is a tradeoff of using smaller tech, unfortunately. Accessibility often gets treated as an afterthought, or maybe more generously, they care in theory but just haven't gotten good at it yet. Big Tech doesn't have that option because they will get sued into oblivion if they aren't accessible.

When I checked back in on this today, the heading structure is even more confusing. It now has a whole bunch of h2's and some h3's properly below them. But none of those headings are actually showing on the dashboard page I'm currently viewing, and there is still no h1. Maybe he's in the process of fixing it directly in production and there's some mess still in place?

It's early. It might get fixed very quickly. It does make me a little nervous, though.

Contacting Users

Speaking of slightly unusual that made me a little nervous, I got an email from the creator Barry. I initially assumed it was a generic automated "welcome" email, but then realized it was personalized, and he had taken the time to look at this tech site. Did I agree to receive personalized email? I truly don't remember. I might have, given the context of a new Canadian effort that seems to have its heart in the right place, and forgotten that there was some checkbox for that. I'll give the benefit of the doubt and assume I did agree in some capacity.

But if I didn't, that is not great. It means he's looking at user emails out of curiosity and then contacting them. That seems less than ideal privacy practice. That's also a bit of a common indicator of smaller tech. Instead of trusting some juggernaut with all kinds of compliance policies and the legal threats if they violate privacy too much, you're trusting... one person who you know nothing about except that he is trying to build a cool thing. On one hand, Big Tech can do way more damage if they violate that trust, but on the other, they also will get into way more trouble if they get caught. Well, at least sometimes they will; there's a lot less digital privacy law in North America than most people realize. I've had this thought lots of times with the admin of my Mastodon server, too. How much do you trust one person or a small group of people that has very little accountability built in?

Backups / Sync

Speaking of trust, what about backups? If I'm working on something for a month, where is my confidence that it will never get lost? It doesn't sync a file to my machine like a Word document in OneDrive, where even if my OneDrive account gets locked out, I still have my local copy. I imagine they do have some redundancy in cloud servers, but it is still putting the trust entirely on them.

Since the developer had already emailed me, I asked him. He said he does hope to have a bulk export tool in the short term. That's nice, and probably a lot easier for him to implement, but does still require the user to remember to take some backups manually after any change they know they don't want to lose. In the long term, he'd like to do a desktop sync client, which is ultimately what I would want but is also a much larger undertaking so I am not holding my breath that it will be any day now.

Pricing

You get 50Mb of documents for free, and there are no other limits on the free plan as far as I can tell (no ads, not even anything intrusive encouraging you to upgrade). If you're just writing the occasional text file for personal use, without many images, that's probably fine. It also seems like a reasonable amount to me to be able to know whether it is useful enough to you to pay for more.

If you need more, it's a standard monthly fee of $8/month, which is reasonable if you are going to use it a lot.

Conclusion

So will I use it? Probably not yet. I'll keep the account and try to remember to check in occasionally. It might improve enough to be worthwhile. For now, though, I'm not sure it's actually addressing any problems I have - the one minor exception for streamlining markdown output - while most likely introducing a few others. I hope he succeeds. I really support the mission of it. It's just not there yet.